Study On Several Key Techniques Of Multi-sensor Battlefield Information Acquisition System Used In Vehicle

This dissertation presented investigates several key techniques that based on sub-task"Multisensor Vehicle Based Battlefield Information Acquisition System"in National 973 project"New Mechanism, New Method Research for Automatic Radar Target Recognition". The first chaper expatiate on background of research and reviewing the developments of object recognition based on data fusion and multisensor management, design of millimeter radar echo signal A/D sampling and processing.In chaper 2, two channels A/D sampling system is designed. Aim at consistency between channels and transfer time synchronization, introduced schemes for error compensate and circuit design. It achieves 220MHz two channels A/D sampling for millimeter radar echo signal. The testing indicates the good performance on SNR, ENOB and channel consistent.In chaper 3, to solve the conflict between great computation quantity and desire for real time processing, base on four ADSP-TS101S, this dissertation design paraller DSP processing system that use combination of shared bus and distributing LINK interface. Then this chaper introduced a frequency digital pulse press arithmetic using pipeline paraller processing.In chaper 4, propose a sensor management method to optimize target recognition based on discrimination. For reference of the algorithm of sensor management based on information entropy, using D-S proof theory, a sensor management measure is realized due to the globally optimal discrimination rule. On the foundation of this, another method of management sensors in fusion target recognition system is proposed. During the process, Bayes rule is used to compute the expected discrimination gain.In chaper 5, the basic theory and applying method of fuzzy neural network in information fusion are addressed here. We discuss the multi-level forward neural network based on fuzzy inference, put forward the training algorithm based on back-propagation, and investigate the general method of applying this model to feature level fusion. Next the general method of applying fuzzy ARTMAP model to feature level fusion is also expounded and we put forward a learning algorithm with adaptive vigilance parameters for each cluster. At last, summary of this dissertation is made and the problems which need further research are pointed out in chapter 6.